Impacts of Surface Roughness Lengths on Typhoon Simulations

Impacts of surface roughness lengths on typhoon simulations are investigated using a nonhydrostatic atmosphere model coupled with an ocean wave model and amulti-layer ocean model with a horizontal grid spacing of 3 km. Five surface-roughness-length schemes used in this study were based on the empirical relation of the length to wave-induced stress, wave age, wave steepness, Charnock constant, and surface wind speed. Five numerical simulations were performed by the coupled model for Typhoon Fanapi in 2009, which is one of targeted typhoons during ITOP campaign observations. The numerical result indicates that the track of Fanapi is well reproduced independent of the choice of surface-roughness-length scheme. Simulated central pressure tends to be higher in all simulations than best-track central pressure archived in Regional Specialized Meteorological Center (RSMC)-Tokyo. Sea-surface cooling induced by Fanapi is reproduced on the right side of simulated track. However, the amplitude of sea-surface cooling is smaller than that derived from a daily Microwave Optimally Interpolated sea-surface temperature. In the future, the coupled model with a horizontal grid spacing of 2 km will be conducted for the purpose of the improvement of the simulations of Fanapi's intensity and sea-surface cooling.

The relation of 10-m wind speeds to drag coefficients indicates that the drag coefficient monotonically increases with an increase in 10-m wind speed except that calculated by the scheme on the basis of wave steepness. The relation calculated by the surface-roughness-length scheme on the basis of wave steepness indicates that the drag coefficient is saturated when the 10-m wind speed is higher than 30 m s^-1.